# ThinkingData MCP for AI Agents AI Agent Connect

> ThinkingData MCP lets you query gaming events, audit user profiles, and manage cohorts using your AI agent. It turns complex analytics into a natural conversation, allowing you to pull data, check schemas, and update properties without jumping through the TA Dashboard.

## Overview
- **Category:** data-analytics
- **Price:** Free
- **Endpoint:** https://edge.vinkius.com/vk_preview_7mJORCNrP18ydJxMazgZnnTvibCa4ej6wVs1Sib6/ai-agent-connect
- **Tags:** behavioral-analytics, event-tracking, user-profiling, cohort-analysis, gaming-insights, data-querying

## Description

ThinkingData MCP lets you manage your game data and player insights through a natural conversation with your agent. Instead of opening a browser, logging into a dashboard, and building custom queries, you can just ask for the data you need. You can pull event data to spot conversion friction, check user profiles to see what high-value players are doing, or even update properties on the fly for testing.

Imagine you are in the middle of a live-ops push and need to know if a specific cohort is hitting a wall. You can ask your agent to pull the latest behavior logs and compare those results against your saved reports to see if the issue is widespread or isolated to a specific region. This replaces the need to manually filter through thousands of rows in a spreadsheet or a complex UI.

You can also use it to track custom events and set user properties directly through the chat. This is a huge time-saver for developers who need to verify that their tracking is working correctly without having to jump back into the codebase or a separate dashboard. When you connect this through the Vinkius catalog, you stop wrestling with menus and start talking directly to your data. It is the fastest way to bridge the gap between your live game data and your daily planning.

## Tools

### query_events
Use query_events to run complex queries on your event data. It helps you understand player behavior and conversion funnels without manual filtering.

### query_users
Use query_users to pull specific profile data for individual users. This is great for auditing specific accounts or checking user status.

### get_event_properties
Use get_event_properties to look up the specific properties tied to a particular event. You can use this to understand the data structure before running a query.

### get_project_summary
Use get_project_summary to get a high-level overview of your ThinkingData project. It provides a quick snapshot of your active events and cohorts.

### list_defined_events
Use list_defined_events to see every event currently set up in your project. This helps you quickly identify which data points are available for querying.

### list_project_cohorts
Use list_project_cohorts to view all the user cohorts you have defined. It helps you check group sizes and membership rules quickly.

### set_user_properties
Use set_user_properties to update the profile properties for specific user IDs. This allows you to modify user tags or status directly through your agent.

### track_custom_event
Use track_custom_event to send a single custom event to the platform. Use this to test new tracking or trigger specific actions during development.

## Prompt Examples

**Prompt:** 
```
What's the current status of our 'High_Spender' cohort?
```

**Response:** 
```
I've checked the cohorts for your project. Here is the summary for **High_Spender**:

*   **Total Users:** 1,240
*   **Active Today:** 312
*   **Avg. Session Length:** 14m 20s

Would you like me to list the most frequent events for this group?
```

**Prompt:** 
```
Show me the properties for the 'Purchase_Complete' event.
```

**Response:** 
```
The **Purchase_Complete** event includes the following properties:

| Property | Type | Description |
| :--- | :--- | :--- |
| `amount` | Float | Total currency spent |
| `currency` | String | ISO code |
| `item_id` | String | Unique product identifier |
| `is_first_purchase` | Boolean | Flag for new buyers |
```

**Prompt:** 
```
Find all users who haven't logged in for 7 days.
```

**Response:** 
```
I've queried the user profiles. I found **42 users** who haven't logged in for 7 days. 

**Sample IDs:**
*   USER_88210934
*   USER_1120394
*   USER_009231

Would you like me to update the 'inactive' property for these users?
```

## Capabilities

### Query behavioral data
Get specific event data to identify where players are dropping off in your game.

### Audit user profiles
Pull detailed profile information for specific users to check their status or history.

### Manage user cohorts
View and organize your defined user groups to better understand different player segments.

### Track custom events
Send new events and properties directly through your agent for rapid testing.

### Get project metadata
Retrieve high-level project summaries and event schemas to understand your data setup.

## Use Cases

### Retention Analysis
A designer asks why players leave level 3. The agent runs query_events and summarizes the drop-off points.

### VIP Auditing
An ops lead needs to see the traits of top spenders. The agent uses query_users to list them.

### Live Event Prep
A dev needs to see if a new cohort is ready. The agent checks list_project_cohorts and summarizes the counts.

### Rapid Prototyping
A developer wants to test a new property. They tell the agent to track_custom_event with dummy data.

## Benefits

- Skip the dashboard: Use query_events to get behavioral data without opening a browser or building manual filters.
- Faster testing: Use track_custom_event to send test data immediately from your agent to verify your tracking setup.
- Precise auditing: Use query_users to pull specific profile details for debugging or checking high-value player status.
- Dynamic updates: Use set_user_properties to modify user tags on the fly without needing to jump into a separate tool.
- Schema clarity: Use get_event_properties to see what data is actually available for an event before you start your analysis.

## How It Works

The bottom line is you get a direct line to your game's analytics without the manual overhead.

1. Subscribe to the ThinkingData MCP on Vinkius.
2. Enter your TA Project APPID, Data Token, Query Secret, and API URL.
3. Start asking your agent for data analysis in Claude, Cursor, or any MCP-compatible client.

## Frequently Asked Questions

**Can ThinkingData MCP help with game retention analysis?**
Yes, it allows your agent to query event data to identify exactly where players are dropping off in your game, helping you spot friction points quickly.

**How do I update user tags using ThinkingData MCP?**
You can ask your agent to update properties for specific user IDs. This lets you modify tags or status directly through your chat interface.

**Can I use ThinkingData MCP to see my project summary?**
Yes, your agent can retrieve a high-level overview of your project, including active events and cohort counts, in one go.

**Is ThinkingData MCP good for live-ops monitoring?**
It is ideal for live-ops because it lets you monitor real-time behavior and verify tracking without leaving your workspace.

**How do I track custom events with ThinkingData MCP?**
You can tell your agent to track a custom event directly. This is great for verifying that your new tracking code is working correctly.

**Does ThinkingData MCP work with Claude or Cursor?**
Yes, it works with Claude, Cursor, and any other MCP-compatible client to give you a natural language interface for your gaming data.

**How do I find my TA Project APPID and Tokens?**
Log in to your [ThinkingAnalytics Console](https://www.thinkingdata.cn/), go to [Project Management] -> [Project Info] to find your APPID and Data Token. The Query Secret is usually found in the [Account Management] or [API Management] section.

**What is the 'distinctId'?**
The `distinctId` is the unique identifier for a user in your system (e.g., a device ID or an account ID). It is used to tie events and properties to a specific individual profile in ThinkingAnalytics.

**Can I perform complex data analysis through the agent?**
Yes! Use the `query_events` tool with a JSON query definition. Your agent will fetch the raw or aggregated results from TA, allowing you to audit behavioral patterns using natural language reasoning over the returned data.